Spaces:
Sleeping
Sleeping
Ahmad-Moiz
commited on
Commit
•
c910b2f
1
Parent(s):
8ffecbb
Update app.py
Browse files
app.py
CHANGED
@@ -2,15 +2,12 @@ import streamlit as st
|
|
2 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
3 |
import torch
|
4 |
|
5 |
-
# Load the tokenizer and model
|
6 |
tokenizer = AutoTokenizer.from_pretrained("mrm8488/distilroberta-finetuned-financial-news-sentiment-analysis")
|
7 |
model = AutoModelForSequenceClassification.from_pretrained("mrm8488/distilroberta-finetuned-financial-news-sentiment-analysis")
|
8 |
|
9 |
-
# Set the model to evaluation mode
|
10 |
model.eval()
|
11 |
|
12 |
def predict_sentiment(text):
|
13 |
-
# Tokenize the input text and prepare it to be used as model input
|
14 |
inputs = tokenizer.encode_plus(
|
15 |
text,
|
16 |
add_special_tokens=True,
|
@@ -19,12 +16,10 @@ def predict_sentiment(text):
|
|
19 |
truncation=True
|
20 |
)
|
21 |
|
22 |
-
# Perform the prediction
|
23 |
with torch.no_grad():
|
24 |
outputs = model(**inputs)
|
25 |
predicted_class = torch.argmax(outputs.logits, dim=1).item()
|
26 |
|
27 |
-
# Determine the sentiment based on the predicted class
|
28 |
if predicted_class == 0:
|
29 |
return "Negative sentiment"
|
30 |
elif predicted_class == 1:
|
@@ -35,7 +30,7 @@ def predict_sentiment(text):
|
|
35 |
# Streamlit interface
|
36 |
st.title("Financial News Sentiment Analysis")
|
37 |
|
38 |
-
user_input = st.text_area("Enter the financial news text:")
|
39 |
if st.button("Analyze"):
|
40 |
if user_input:
|
41 |
# Get the sentiment prediction
|
|
|
2 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
3 |
import torch
|
4 |
|
|
|
5 |
tokenizer = AutoTokenizer.from_pretrained("mrm8488/distilroberta-finetuned-financial-news-sentiment-analysis")
|
6 |
model = AutoModelForSequenceClassification.from_pretrained("mrm8488/distilroberta-finetuned-financial-news-sentiment-analysis")
|
7 |
|
|
|
8 |
model.eval()
|
9 |
|
10 |
def predict_sentiment(text):
|
|
|
11 |
inputs = tokenizer.encode_plus(
|
12 |
text,
|
13 |
add_special_tokens=True,
|
|
|
16 |
truncation=True
|
17 |
)
|
18 |
|
|
|
19 |
with torch.no_grad():
|
20 |
outputs = model(**inputs)
|
21 |
predicted_class = torch.argmax(outputs.logits, dim=1).item()
|
22 |
|
|
|
23 |
if predicted_class == 0:
|
24 |
return "Negative sentiment"
|
25 |
elif predicted_class == 1:
|
|
|
30 |
# Streamlit interface
|
31 |
st.title("Financial News Sentiment Analysis")
|
32 |
|
33 |
+
user_input = st.text_area("Made by Ahmad Moiz with Love","Enter the financial news text:")
|
34 |
if st.button("Analyze"):
|
35 |
if user_input:
|
36 |
# Get the sentiment prediction
|